Metadata-Version: 2.4
Name: fidelitybench
Version: 0.0.1
Summary: A communication fidelity benchmark for multi-agent AI systems. Measuring per-hop information degradation in LLM agent chains.
Author-email: Varun Pratap Bhardwaj <qualixar@varunpratap.com>
License-Expression: MIT
Project-URL: Homepage, https://qualixar.com
Project-URL: Repository, https://github.com/qualixar/fidelitybench
Keywords: ai-agents,benchmark,multi-agent-systems,communication-fidelity,qualixar
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# FidelityBench
A communication fidelity benchmark for multi-agent AI systems.
Part of [Qualixar](https://qualixar.com) | Author: Varun Pratap Bhardwaj
**Coming soon.** Follow [github.com/qualixar/fidelitybench](https://github.com/qualixar/fidelitybench) for updates.
## License
MIT
